The parts that go first, and what that costs a plate
This is not a minor technical problem. Soft parts are identity. A laughing gull in winter plumage and a laughing gull in summer plumage share the same feather geometry; what distinguishes them at a glance is bill colour and hood. A cattle egret in breeding condition declares itself through a flush of orange-buff on the crown and, critically, through the brief livid red of its loral skin — orbital skin that will have greyed and sunken before the bird is dry. An artist who has only the specimen has already missed it.
What survives, what does not, and what gets invented
Feather pigments based on melanins — the blacks, rusts and buffs — are chemically stable and survive well. The structural colours produced by feather microstructure, the iridescent greens and blues of a mallard's speculum or a hummingbird's gorget, persist too, since they are physics rather than chemistry. Carotenoid pigments, responsible for the yellows and reds of soft parts and some plumage, are considerably less stable; the vivid red of a recently dead finch's bill begins to dull and dry within hours.
The eye is the most acute problem. A living eye is a wet optical instrument: it has depth, reflection and a catch-light that locates the animal in a real moment of attention. Within an hour of death the eye goes flat, the cornea clouds, and the rounded geometry of the globe begins to collapse. What the artist then paints is a disc rather than a sphere, and the bird reads as stuffed regardless of every other virtue in the picture.
This is why field notes on colour are the primary document, not preliminary sketches. A careful field observer recording a wader at the water's edge will note the exact khaki-olive of the tarsus, the extent of the red at the bill base, the yellow of the orbital ring against the dark loral stripe. These notes can be taken in minutes. They do not require a long observation or a still subject. What they require is that the observer knows to take them — knows what the specimen will not preserve.
When those notes do not exist, the plates show it. In a number of eighteenth- and early nineteenth-century illustration projects, the bills and legs of waders are painted in colours that correspond neither to the living bird nor to any known subspecies: they are the artist's reasonable interpolation from a dried or preserved specimen, which is to say they are quietly wrong. The systematic distortions introduced by preservation rather than ignorance are consistent enough across plates to constitute a visual fingerprint: legs a little too brown, bills too dark at the tip, eyes too dull and too small.
The plates made from living birds, or from freshly killed birds with field notes immediately to hand, read differently.
The plates made from living birds, or from freshly killed birds with field notes immediately to hand, read differently. There is a specificity to the soft-part rendering — the exact point at which a bill changes from one colour to another, the slight difference in tone between the front and back of a tarsus — that cannot be guessed from a dried skin and cannot be invented with any conviction. The difference, once you have learned to look for it, is not subtle. One painting is of a bird. The other is of an artist's idea of what a bird must look like, assembled from parts that had already begun, quietly and irreversibly, to leave.
